A transactional read-response application is used when you need to transactionally retrieve messages from a remote transaction queue. It uses transactions to redirect the messages from the remote queues on the sending computer to local queues on the receiving computer.
In this scenario, the read-response application resides on a server that contains two transactional queues: an input queue for collecting messages from sending applications, and a read-request queue for collecting read-requests from the receiving applications.
The read-response application uses a single transaction to retrieve a request from the request queue, retrieve a message from the input queue, then forward the message from the input queue to the receiving application that made the request. In this way the receiving application gets its messages within a transaction without the need to perform a transactional remote-read operation.
To get its messages, the receiving application sends a request message to the server specifying the response queue to which the input message should be sent to. This request can be made within the context of its own transaction. However, it is independent of the transaction used to send the message back to the receiving application.
More Information
| For information on | See |
|---|---|
|
Restrictions on sending or retrieving messages within a transaction | |
|
A summary of the different transaction types provided by Message Queuing | |
|
Example code of sending message in different transactions |